Arif Hasan Q&A

Justthetippett · May 9, 2024

Thanks for pulling this all together @SMU_Sox @Dogman.

Short question from me @ArifMHasan. I would be very interested to get your overall assessment of drafting for need vs. BPA. These tend to be presented as mutually exclusive choices, and I can't imagine teams think of things that way. Teams both need good players, and they can only carry so many at each position to build what Belichick referred to as the team "mosaic". Is this what leads teams to put together tiers of players, such that they have similar grades irrespective of position? How is need balanced with trying to maximize the value of a particular pick (and how do we define value)? I would imagine the BPA thing also weighs differently depending on the round. For example, the Pats may not have needed a DT in the Wilfork draft but he was such an underrated player in the back half of the first round, that they didn't pass him up. Maybe this is different when you get to rounds 3-7, where the success rate is lower and variance is so much higher. Thanks!

Cellar-Door · May 9, 2024

So I was reading through a bunch of wide-left (very interesting stuff)....

One question I had is... how much do you think you are getting diminishing returns the further down the board you go?

In your analysis of past drafts you noted that a lot of the boards that go into the consensus only go 100 deep, which would seem to mean that once you get much if any past that you are putting a lot more weight on a smaller number of boards.

In your analysis of hits/misses with regards to reaches/steals, it also looked like once you got past the 2nd round reaches didn't really seem to mean much (you didn't have anything broken down beyond top 100).

Is there any thought to whether this is just a limitation of the boards in general that the top 50 to 100 is bringing some real value, but beyond that for whatever reason (bigger impact of scheme/fit on lesser prospects, less ability to project, etc.) the boards aren't bringing the same value?

Ale Xander · May 9, 2024

Will there be a V&N Q&A given Arif’s expertise?

Ale Xander · May 9, 2024

Politics only on V&N per mod rules.

Hard to answer your q

Norm loves Vera · May 9, 2024

@ArifMHasan I am curious about your interactions with college football coaches when vetting players from their team. It seems so competative do coaches try to pitch their athlete to you to get a higher ranking or more buzz from you?

Dogman · May 9, 2024

Absolutely no politics outside of V&N. None. Zip. Zilch.

No leeway whatsoever on this.

Please stick to Arif's Football expertise in this forum.

Thanks.

j-man · May 9, 2024

i am j-man on mile high report what do u think about the broncos off-season

is nix a mac jones type or better
can denver stop the run better with franklin-myers can Eillis grow into a starter 3rd round pick 76 overall
assuming its a lean next 2 seasons what can denver do to colpete for a wild card spot by 2026

nice to see u do so well for yourseif

Dotrat · May 9, 2024

Thanks so much for setting this up, @SMU_Sox--and than you @ArifMHasan for joining.

Making fun of various teams' drafts is a cottage industry among fans and media (with the Falcons winning first prize in 2024's 'WTF were they thinking?" sweepstakes). Yet year after year we also see how wrong so many draft 'experts' were with regard to everything from a player's ability to system fit. Given that you came up in around analytics, I'd like to know which factors play the biggest role(s) in how you weigh a player's potential. And what lessons have you gleaned from your work that you'd like to see used by more teams in evaluating players?

Garshaparra · May 9, 2024

Hi Arif! I am a pro sports fan, eschewing viewing of the college games for a variety of reasons. I find myself fascinated though with the ever-changing structures of college football, with the Power 5 coalescing into a Power 4 (or perhaps Power 2), the constant roster turnover from the transfer portal, and NIL/agency for college-age athletes. How has that churn impacted the ability to scout player effectiveness as they move to the pros?

AlNipper49 · May 10, 2024

Hi Aarif,

Thabk you for coming here to talk to us.

I have a more basic question than most I presume. I have a 16 year old kid who is obsessed with following the NFL. It’s pretty clear you’ve taken a systemic approach to learning about the game. Other than your blog (which he’s not getting updates for) what other resources would you suggest for folks who want to follow the NFL and grow the knowledge of the game?

Basically, somewhere below becoming a Dan Hatman student and above an average WEEI listener

Ale Xander · May 10, 2024

Hi Arif,
Thanks for doing this. Do you think the NFL will be around in 30 years? Why or why not? What is the greatest obstacle, CTE awareness, greed, or something else?

jacklamabe65 · May 10, 2024

Welcome, Arif!

Do you think the Patriots should sit down with Blake Maye and have him learn the ropes this year (as long as Jacoby remains healthy)?

Eck'sSneakyCheese · May 10, 2024

Welcome Arif! I think it's great that you're taking time out of your obviously busy schedule to lend some of your expertise to our conversations. I look forward to your replies on some great questions already brought up by others. I read a couple of the articles on your site and they were really informative.

I'm really interested in the draft evaluation process but unlike most here I'm leery of how quantitative something like this truly is. There are numerous variables at play that can't be measured. The human element and it's uncertainty is by far the biggest one. Knowing that there are multiple environmental attributes including the ever evolving landscape of the NFL, it's rules and it's personnel that can effect outcomes how can any of this number based analysis be trusted? Furthermore, taking into account the fact that there are so many people that have opinions on prospects now and some are more clued in than others are you worried about having so many boards creating noise in the final tallies?

sezwho · May 10, 2024

Thanks for joining! My question is a random one about dline play. With the continued trend towards multiple defensive fronts is it realistic/effective for a d tackle to go back and forth between 1 and 2 gap techniques even in the same series?

dirtynine · May 10, 2024

Hi & thanks for dropping by.

Whenever I try to watch all-22 or even YouTube breakdowns, I get lost. Too many tiny moving pieces, not enough inherent understanding of what any particular offensive play is “supposed” to be or what the defense is “trying” to do. When you’re watching film, are there any basic pointers you’d give to inexperienced viewers that can lend a frame to things? A process or checklist to use before the snap to set up a mental model for what you’re about to see?

Bowser · May 10, 2024

Thanks, Arif, for spreading some wisdom.

Question: How seriously should we take any consensus board?

I assume we're referring to the consensus of talking head analysts and scouting services, i.e., not the teams themselves? Given that each team has a better understanding of its needs and how it values the various position groups, is there any real benefit in, say, critiquing the Pats' recent draft performance by comparing its internal assessment of needs (which we can't know) to some consensus of non-decision-makers?

Awesome Fossum · May 10, 2024

AlNipper49 said:
Hi Aarif,

Thabk you for coming here to talk to us.

I have a more basic question than most I presume. I have a 16 year old kid who is obsessed with following the NFL. It’s pretty clear you’ve taken a systemic approach to learning about the game. Other than your blog (which he’s not getting updates for) what other resources would you suggest for folks who want to follow the NFL and grow the knowledge of the game?

Basically, somewhere below becoming a Dan Hatman student and above an average WEEI listener

Oh great question. On a sort of similar theme: How do you watch football on TV? That is, when you sit down to watch a single game live on TV, what are you paying attention to pre-snap and post-snap? Are you just following the ball or something else? Are you taking mental notes? Physical notes!? Is there anything you'd recommend for people who want to be able to follow a given given at just a slightly more advanced level?

brendan f · May 11, 2024

Hi Arif,

Thank you for taking the time. In the draft, it was widely reported the Patriots had thought they were going to move up and draft either Legette or Coleman (ranked 33, and 36, respectively) at the end of the first round. Instead, the trade fell through and they ended up with Polk and Baker, (ranked 62, 95, respectively).

While we cannot say with certainty that had the trade been successful it would have precluded them from drafting Polk and Baker, for the sake of this exercise, how would you weigh the opportunity to move up in this scenario and draft the consensus 33-36th ranked player versus trading back a few spots and drafting both the 62nd and 95th ranked players?

ArifMHasan · May 12, 2024

Re-posting a question from the other thread and my reply:

EL Jeffe said:
Oh wow, this is really cool! Thank you to Arif for lending your time and expertise. And thank you to John and everyone else who played a part in getting you here!

I have a million questions and could pick your brain all day, but I'll start with 2 if that's okay?

How do draft analysts account for someone like Giovanni Manu when they create and maintain their big boards? As far as I can tell, he wasn't on the radar until his measurables came out at his 3/29 B.C. pro day and he ends up being a 4th round pick (#126) a month later. I imagine NFL teams had access to his Canadian all-22 film and he was scouted live during the season, but I don't know how readily available his film would be for the draft analysts out there. When there's such limited information about a player but his measurables are extremely rare at a scarce position, what's a reasonable way to rank that player? (It makes me wonder how draft analysts would have rated a Matt Cassel or even an Eric Swann back in the day before online draft analyses exploded in such a meteoric way.)

Thanks again!

I'll be honest, I don't think that the current draft media environment does a great job with international prospects. While I suspect the NFL slightly overvalues them (and had even before Jordan Mailata succeeded) I think that analysts undervalue them to a far greater degree. I think this will eventually change but for now I think it's just not having these players on their radar. Film may not be easily available for Canadian football players, which obviously matters, but there are workarounds for that. For other international players, it's a bit different. Some played football (like Moritz Boehringer and Babatunde Aiyegbusi) and some did not (Jordan Mailata and Jarryd Hayne). The ones that did not just have to be incorporated as if they are pure athletes (the same way that some basketball players or track athletes get valued). It obviously would be up to each analyst how they want to evaluate pure athletes, but they should at least be open to them. Someone like Dan Chisena was still valuable for the Vikings despite playing about 20 snaps of football in college because of his pure track speed — because there was a special teams role for him. They would probably want to build a checklist for these kinds of players ("Have we seen them engage with physicality or contact? How do they react to it," "Are their movement patterns linear or dynamic?" "Do they have to handle a ball, and how did they do so?" etc) and slot them based on the answers to those questions along with athletic testing data.

This undervaluation of international prospects (and specialists) is one reason why I produced two draft grades this year -- one that used consensus big board data and compared it to every pick and one that excluded picks for international players and specialists.

EL Jeffe said:
When an NFL "insider" like Tom Pelissero tweets out: "A name to remember in next week’s draft: Penn State OL Caedan Wallace. He had nine “30” visits with teams, including six in the last 10 days. A potential Day 2 pick" a week before the draft, and he's ranked in the 180s on the consensus big board, is that something draft analysts do (or should) take into account? Do they ask "what is the NFL seeing that I'm not seeing?" Or do they stick to their proverbial guns and trust their evaluations? I'm sure it varies on the analyst but I wonder how much attention is paid to that type of insider reporting. When there is that sort of disconnect between the consensus big board and how at least some NFL teams view a player, it makes me question how or why that disconnect happens. Maybe OL coaches fall in love with a guy based on private workouts that analysts don't have access to and it becomes an artificial rise? Or is it a case where too many analysts just mis-evaluated the player for whatever reason? I know some like Thorn and Brugler had him more a 4th rounder (making him a reach but not a massive reach), but outside of the Pelissero tweet, I'm not aware of any analysts thinking he could go as high as he did.

Thanks again!

The data suggests that visits provide signal to eventual draft position beyond mock draft and big board data. Another way to put it -- any model that attempts to predict draft position will improve with the addition of visit data (presumably when excluding locals, which these reports sometimes do not do -- if Caedan Wallace visited the New York Giants, New York Jets and the Philadelphia Eagles, I probably would not count that). Nine visits is quite a bit for a player ranked outside of the Top 150 on the consensus board. Joe Milton, ranked right next to Wallace on the Consensus Board but at a much more valuable position, only had two visits! Walter Rouse, ranked next to Wallace but at the same position, seemingly had zero visits.

What I have not seen tested is whether or not a model that attempts to predict player outcomes improves with the input of team visits. My intuition is that it would be given how useful draft position is to predicting player outcomes, with or without the board. Maybe I should look into it a bit more -- do reaches who have had a lot of team visits outperform reaches without those team visits? If I did that, I would also want to compare that to the data produced by those who produce anonymous scout quotes on teams, because those are also revealed (if more selective/biased) preferences. Bob McGinn's survey of NFL personnel, for example, did not regard Wallace as one of the top 12 tackles in the draft. At least one other scout told Mike Giardi (you know more about his credibility than I do) that the pick didn't make much sense.

As for what analysts should do, it really depends on their own internal process. One thing I really like about this project is seeing the different approaches different analysts take. Some purport to only use film analysis, at the exclusion of production data, athletic workout data, off-field concerns (including arrests), press appearances/interviews and rumors. Some will include all or some of those factors in their evaluations. Some of these boards have different goals -- some will want to be a lot closer to the NFL while others want to be independent of the NFL. Predicting draft order will matter a different amount to different analysts. Were I to have the time and energy to put together my own big board, I would use high-visit count as a re-check flag and go back to re-watch any prospects that ranked low on my board but had a number of team visits.

It is perhaps relevant that two sources that have a bit of a better bead on what NFL teams think -- Lance Zierlein and the person now running Scouts, Inc (Steve Muench) were the highest and third-highest on Wallace, ranking him 80th and 106th respectively. Others who have some insider access include Jeff Legwold, who was close, ranking him 111th while Field Yates ranked him 114th. That's one reason the Forecaster Board (composed of people with more insider access) ranked him 157th while the Evaluator Board ranked him 191st.

Generally speaking, reaches have a much, much worse track record -- even outside of the Top 100 -- than even picks and steals. Substantial reaches that were nevertheless ranked much higher by forecasters still carry that poor track record. But were I to reach for a player it would be for a player with a high variance score (Wallace's variance was about average for his rank), who was identified by NFL executives as a potential sleeper (e.g. Darius Robinson, Mike Sainristil or Quinyon Mitchell) and for a player who took a number of team visits. Wallace only meets one of those three criteria to my knowledge, which suggests to me that his Consensus Board ranking is a little low on him but not extraordinarily so.

ArifMHasan · May 12, 2024

Justthetippett said:
I would be very interested to get your overall assessment of drafting for need vs. BPA. These tend to be presented as mutually exclusive choices, and I can't imagine teams think of things that way. Teams both need good players, and they can only carry so many at each position to build what Belichick referred to as the team "mosaic". Is this what leads teams to put together tiers of players, such that they have similar grades irrespective of position? How is need balanced with trying to maximize the value of a particular pick (and how do we define value)? I would imagine the BPA thing also weighs differently depending on the round. For example, the Pats may not have needed a DT in the Wilfork draft but he was such an underrated player in the back half of the first round, that they didn't pass him up. Maybe this is different when you get to rounds 3-7, where the success rate is lower and variance is so much higher. Thanks!

My thought process primarily revolves around the 4,800 snaps you have available to you when you draft a player. If it's a first-round player than can increase to 6,000 but the point is that the rookie contract represents those snaps and if you can maximize the value of those snaps, you should. That generally means finding a player who will play well, play often and do so at a valuable position. It also means that the player they are replacing will be necessarily that much worse. If you replace a Pro Bowler with an All-Pro that's good but generally not nearly as good as trading a borderline starter for a high-quality starter. Obviously, this is a bit too simple -- rookie contracts give you an exclusive negotiating window if that player turns out to be good, but you will essentially be giving up the advantage of having that player when they actually realize their market value (which is a screwed up system!).

The board is of course geared towards BPA but from the perspective of an individual team I tend to bias towards need. Of course, if you have very few needs and none of the players at those positions are ranked close to where you are, then you should either draft BPA (with caveats) or trade down. But, generally speaking, I think you should maximize the potential available snaps of your young players. A wonderful pick does very little on the bench. And even if your excellent BPA pick is better than your high-level starter and replaces him, the marginal value you get out of that is pretty small. How much would you spend to upgrade from Tyler Lockett to Amon-Ra St. Brown if it meant benching Lockett? Probably a fair amount, but not enough for you to meet the fair market price for St. Brown and carry Lockett's contract. We often don't see draft picks that way -- if a player is good, he's good! But it's good to think about how an action now changes your actions in the future.

The Vince Wilfork example is an interesting one. He was generally regarded as a top-ten player in that draft (if I recall correctly) so falling to to 21 is nuts. Also, are we confident this doesn't qualify as a need? Keith Traylor was 35 years old on a one-year contract.

The caveat does come at that position, though -- there are positions where you benefit from rotation or additional players and positions where you don't. Generally speaking, you don't see much change in cornerback, safety or offensive line. You generally see more rotation at edge rusher, defensive tackle and kind of at receiver. I say kind of because I think you can max that out. For example, Seattle isn't getting much out of Jaxon Smith-Njigba despite his talent at the position and would have been better off drafting Brian Branch last year to play safety (or nickel, leaving it as Riq and Devon on the outside) and drafting a WR this year or next to prepare for Lockett's eventual departure. Even if JSN is good I don't think that invalidates that point -- someone else they may have drafted in a later year could also be good. Unless Lockett had already departed this year, I don't think this is that analogous to the Wilfork example.

But if there is no possibility of a trade and no possibility of avoiding a reach by several tiers, I'd rather select BPA than draft for need -- if I am picking 12th overall and my needs are WR, OL, and CB and the only players ranked lower than, say 22nd, were DTs, EDs and LBs I'd probably take the edge rusher (or a falling steal) and try again at one of my needs later. But to me that's a bit of an extreme. I think the differences between ranks at a granular level tend to be pretty small, so early in the first round I'd be comfortable reach two or three spots for need and in the middle of the first round six or seven spots.

Later in the draft, the differences in players are very small but so are the odds that this player will see the field soon. I would rather maximize the odds that they are a good depth player by drafting BPA later on, primarily by targeting steals or players where I think I have a unique deployment advantage (a one-dimensional player who is a scheme fit) or unique scouting advantage (e.g. an international player).

ArifMHasan · May 12, 2024

Cellar-Door said:
So I was reading through a bunch of wide-left (very interesting stuff)....

One question I had is... how much do you think you are getting diminishing returns the further down the board you go?

In your analysis of past drafts you noted that a lot of the boards that go into the consensus only go 100 deep, which would seem to mean that once you get much if any past that you are putting a lot more weight on a smaller number of boards.

In your analysis of hits/misses with regards to reaches/steals, it also looked like once you got past the 2nd round reaches didn't really seem to mean much (you didn't have anything broken down beyond top 100).

Is there any thought to whether this is just a limitation of the boards in general that the top 50 to 100 is bringing some real value, but beyond that for whatever reason (bigger impact of scheme/fit on lesser prospects, less ability to project, etc.) the boards aren't bringing the same value?

I looked at the average number of rankings a player receives by rank going down the boards and the reason I stop at 300 for the Consensus Big Board is because I still have about 40 boards with a ranking on a player at that point (300th-ranked Jaden Shirden from Monmouth received a ranking from 39 boards this year). One thing to keep in mind is that even though the minimum requirement for a board to be included is to rank 100 players, they are not ranking the same 100 players. The first player not on all 101 boards is 37th-ranked Edgerrin Cooper, followed by 40th-ranked Ennis Rakestraw Jr. Whenever something like that happens and I have a line of communication to the analyst, I double-check to make sure this was intentional. Often it is, sometimes it isn't.

100th-ranked Khyree Jackson was ranked by 83 boards while 101st-ranked Malik Washington was ranked by 82 boards. Because of this, I do have a fairly robust ranking for most players going deep into the board. It drops off fairly quickly after 300 (the first player with fewer than 20 boards is 357th-ranked Zakhari Franklin), which is why I typically don't go that much deeper than 300 on initial publish.

With that in mind, I do think there's a fair bit of fidelity deeper in the draft and the areas where the Consensus Board struggles are already mentioned above -- international players and specialists. OverTheCap found that for players ranked 128th or later get larger second contracts than those picked 128th or later in the NFL draft, when parsing it out into buckets of picks. This suggests that there is something concrete about steals. PFF did find that the NFL was a bit better than the Consensus Big Board using their own methodology on those late-round picks but to my eye they are astonishingly close. I've found using my own methodology that for late picks, steals outperform reaches by a small but statistically significant amount. What's interesting is that in the first 100 picks, the NFL outperforms consensus on even picks (e.g. picks where the value of the pick and the value of the consensus rank are with 15%) but after pick 100 the Consensus outperforms the NFL on even picks. The NFL outperforms on both steals and reaches in that pick range, but the steals still outperform the reaches.

What all of this tells me is that it's probably the case that they are very close. One method (OTC) found the CBB outperforming the NFL late. Another (PFF) found that it was about even with a slight NFL advantage. And my own found that the NFL had a larger advantage.

I largely focused on the Top 100 because I didn't have an enormous amount of time to write the piece and also the last time I isolated late-round steals, people would argue endlessly over whether or not someone was a steal. Dak Prescott and Amon-Ra St. Brown were inarguably wins for the Consensus Big Board but it was very obnoxious making the case that a seventh-round pick seeing playing time at all is a win because people do not want to hear that Myles Gaskin is a win for the Consensus Big Board. No one is interested in the idea that because Kelvin Harmon was much worse than his CBB ranking that he still outperformed his pick slot of 206th.

It should also be noted that the CBB can identify a steal and be correct that this player would outperform their draft position but that the NFL was closer. That's the case for players like Maurice Hurst (the NFL was far, far closer to his value than the CBB... but it was correct to call him a steal) and Jalen Mills.

I tend to focus on the Top 100 because that's who people read about and they are also the players with less ambiguity surrounding them.

ArifMHasan · May 12, 2024

Norm loves Vera said:
@ArifMHasan I am curious about your interactions with college football coaches when vetting players from their team. It seems so competative do coaches try to pitch their athlete to you to get a higher ranking or more buzz from you?

I don't often interact with college coaches (and usually just the analytics or strength staff when I do) but honestly we mostly just talk shop; I rarely ask about players. For example, I've had discussions about a cognitive test that was being used at the college level that I was suspicious of (not the one that made the news a bunch last year) and we talked through the challenges. I have talked to a recruiter about projecting size changes for high school athletes (this guy ran a 4.50 at 190 but we think he has the frame to be a tight end, what should we project him to run if he adds 60 pounds?) and so on. But I have a lot of colleagues who more directly interface with coaches at the level you're talking about and I can speak to the third-hand experiences I've gathered from that.

In my third-hand experience, coaches want to talk about players who they think are either underrated or overrated, typically the former. It's not usually about a third-round guy that they think should be a first but usually about a player who might not be considered a priority free agent that should be on a UDFA board. But coaches do like to be honest -- if they think a player is limited, they will often tell you. I tend to think they still overrate their own players but that's normal. The thing to keep in mind is that getting players drafted helps them, sure, but that's not their priority. Adding a point to their recruiting pitch is fine, but they often want to find ways to maximize next year (or if you're talking to them before or during a season, that year).

What that means is if you are canvassing the country for players to spotlight, coaches may feed your their second- or third-best player because they don't think their best player (or most appropriate player for whatever list you're compiling) is putting in the work or has a too-big head. But yes, there are instances where coaches will try to talk up players to get more buzz, it's just more complicated than that.

One thing I always find cool is when you ask coaches about in-conference players on other teams.

ArifMHasan · May 12, 2024

I'll get to more of these later! A lot of great questions

Manuel Aristides · May 13, 2024

Hey Arif, thanks for being here.

One thing that I end up debating a lot with fellow football junkies is the health and dependability of running quarterbacks. Anecdotally, it feels easy to say that these guys get hurt more often. Richardson's rookie year felt like the archetypal running QB performance: inspired, game breaking talent when he's out there, but the same things that make him undefendable seem to increase his injury risk. While I personally feel like this is a "thing", some of my more data savvy friends have crunched numbers and found other numbers crunched by others and apparently the math is in opposition to my observations. Which, sure, I can believe I might be wrong: even just last year, significant injuries accrued even to obvious pocket-guys on all levels of the talent spectrum; Herbert, Cousins, Jimmy G, just off the top of my head.

The implications are obvious for both the fantasy obsessed (yo) and the actual gridiron. Teams seem to think there's not much to my fear, as running QBs continue to rise in drafts generally. Would love to know if you think there's an actual answer to the question "Is building around a running QB long term a bad idea, because they're more likely to be injured (or for some other reason not listed here, ie, maybe those guys never hit their ceiling as a passer or whatever)"?

Ideally would love your answer before my dynasty draft next week. (Kidding, mostly.)

ArifMHasan · Wednesday at 3:02 PM

j-man said:
i am j-man on mile high report what do u think about the broncos off-season

is nix a mac jones type or better

I love Mile High Report!

I'm pretty down on Nix, generally, and thought more of Mac Jones coming out of college than I do Bo Nix. I was obviously wrong on Jones and there's a good chance now that Nix is a better quarterback but if I'm sticking only to pre-draft evaluations, I think there's more to be concerned about.

With Nix, I'm uncomfortable with his high rate of throws short or behind the line of scrimmage, but am enthused about his arm, which is stronger than Jones'. I think choosing to simply isolate the plays where he throws intermediate or deep is a flawed approach because the defensive structure is crowded to the line of scrimmage, even on third and long, and so the onus on him is still relatively small and non-transferrable. The fact that he didn't really take off until he transferred to Oregon deepens this concern. He's not the only one where this happened -- Michael Penix and Jayden Daniels both succeeded primarily after transferring -- but Penix wasn't actually bad at Indiana, at least in 2019. And Daniels apparently had captured the attention of scouts in a very positive way at Arizona State despite his mediocre numbers.

Succeeding after transferring is not a red flag to me, but it magnifies an already existing red flag, in my opinion. I do have system concerns about Penix and Daniels, but they aren't nearly as dramatic as my system concerns for Nix. I don't see his ball placement on intermediate or deep throws as anything special in particular and I think he declines options on standard reads far too often. I think this can be mistaken for quick processing but I think it's evidence of a lack of processing -- not anticipating defensive coverage well enough to demonstrate patience and make a play. He only had to go past his first read 10 percent of the time! Only 9.5 percent of his targets went into tight windows. On that small percentage of tight-window throws, his accuracy rate was in the middle of the pack.

I think Mac Jones is not quite the same kind of prospect; he was a processor who went through reads fairly well but didn't have the arm strength or accuracy to make good on his profile.

j-man said:
can denver stop the run better with franklin-myers can Eillis grow into a starter 3rd round pick 76 overall

As for the defensive line questions -- Franklin-Myers is an underrated, extremely talented player who is big as hell for an edge rusher that uses his strength to beat back tackles and shed later to get pressure or get after the run. I believe Denver will play him on the inside, which should be his best spot in that defense. I think Denver's run defense will improve with the addition of Franklin-Myers, for sure. He's not Danielle Hunter or Jonathan Greenard when it comes to run defense. But he's better than average and he's consistent.

Jonah Elliss is secretly one of the most athletic players at his position in the draft. His get-off distance in the first second of the snap is up there with James Pearce Jr. (a candidate for the first overall pick next year), Princely Umanmielen and Chris Braswell (seventh on Bruce Feldman's Freak List), in the 99th percentile of the statistic. It's not just that Elliss tested well (he did, of course) but it shows on the field. That he had an excellent explosion score (38-inch vertical) and an astounding agility score (6.69 seconds in the three-cone) only helps -- these correlate strongly to edge rusher production. And he produced -- 85th percentile or better in stable pressure production statistics. He doesn't have length or height, but I think he's just good. I think he'll make an impact year one (and could very well start bc of who is in front of him) and could be a starting-quality player (regardless of if he actually starts or has started) by year two.

j-man said:
assuming its a lean next 2 seasons what can denver do to colpete for a wild card spot by 2026

Denver can hope I'm very wrong on Bo Nix, I guess. Other than that, they've managed their offensive line well and have a good crop of receivers. The defense has a great foundation with Pat Surtain and some of their defensive linemen but I think what they really need to do to make 2026 possible is invest heavily in that secondary to support Surtain. It's there, but there's a lot of work to do.

ArifMHasan · Wednesday at 3:22 PM

Dotrat said:
Thanks so much for setting this up, @SMU_Sox--and than you @ArifMHasan for joining.

Making fun of various teams' drafts is a cottage industry among fans and media (with the Falcons winning first prize in 2024's 'WTF were they thinking?" sweepstakes). Yet year after year we also see how wrong so many draft 'experts' were with regard to everything from a player's ability to system fit. Given that you came up in around analytics, I'd like to know which factors play the biggest role(s) in how you weigh a player's potential. And what lessons have you gleaned from your work that you'd like to see used by more teams in evaluating players?

Production and athletic data are great, but the most useful datapoint I've found has been subjective scouting reports. There are a lot of ways to deploy them -- I use the Consensus Big Board of course, but I've seen a lot of people do other things; Eric Eager was working on semantic analysis -- which pulled out keywords and sentiment indicators of scouting reports (the test sample was on Dane Brugler, but can be applied to any set of scouting reports) to enable some surprisingly accurate projections. He was just hired by the Panthers. I've also seen factor analysis (called PCA) that groups non-numerical traits (say stuff like "strong arm," "good anticipation," etc) to put players into profile buckets. So, to me, I think the production and athletic data are very important but I've still never found anything beat traits-based analysis.

What I find really funny is how many dynasty fantasy experts dismiss subjective scouting but craft models where the most powerful input is player draft position. Like... that player was selected there primarily as a product of subjective scouting.

The thing I've found is that when you find an edge in the data, teams catch up pretty quickly. Justis Mosqueda discovered a model developed by a Football's Future forum user named Waldo to predict edge rusher production and he refined it. Based off of their behavior, it seemed like very few teams were using a similar model, except for the Minnesota Vikings (which is funny, given that Mosqueda is a Packers blogger). That model allowed them to secure really high-level mid-round picks like Ray Edwards, Brian Robison, Everson Griffen and Danielle Hunter. But after 2015, players who would be hits from this model started going earlier and earlier in the draft. Now there's virtually no analytical edge in that regard.

Lessons I've learned: athleticism matters more at some positions (edge rusher, undertackle) than at others (nose tackle, quarterback); test as many things as possible (receiver models work better if you look at ten-yard splits or flying- 20s instead of the 40-yard dash); be aware of changes in player deployment (long corners are statistically not as valuable in a league with less Cover-3); ask questions.

Two ones that stick with me: 1) contrary to most things on the internet, where negativity sells, people will strongly tend to remember your hits than misses. Call shots, be proud of yourself publicly while acknowledging the possibility of being wrong. 2) You're always wrong.

ArifMHasan · Wednesday at 3:37 PM

Garshaparra said:
Hi Arif! I am a pro sports fan, eschewing viewing of the college games for a variety of reasons. I find myself fascinated though with the ever-changing structures of college football, with the Power 5 coalescing into a Power 4 (or perhaps Power 2), the constant roster turnover from the transfer portal, and NIL/agency for college-age athletes. How has that churn impacted the ability to scout player effectiveness as they move to the pros?

I am also primarily focused on professional ball as well, so I can't speak with 100 percent certainty here, but I think the conference change is going to frustrate the traditional scouting model but not by as much as you'd imagine. Scouts are more regional than they are conference-dependent. The West Coast scout will still be trawling through the West Coast and watching games, but the fact that they will be less familiar with opponents (if you watch UCLA three times because you've scouted Arizona, Oregon and Washington, then watching them against USC gives you a lot of context. Not so when Purdue comes to visit) will add some difficulties to their job. They will have to be a bit more aware of the way colleges schedule games, especially if some colleges choose to alter their normal weekly schedule to deal with longer travel distances.

NIL/agency will impact things but for an NFL team, it could be a blessing in disguise. While they will have to be much more comfortable with older prospects (players are no longer desperate to leave college because they now get paid, players can improve their draft stock by transferring, etc) they will get some data that they think will be useful (and they could be correct). If you've read the recent Athletic story on NIL but how players interact with coaches and administrators is kind of fascinating. If they start their transfer visit by talking money first, that could tell us something about their personality (not necessarily bad!). How they treat agents (some players secure an agent, strike a deal and immediately drop that agent for another one to get a new deal), administrators, etc. could be interesting.

Increased transfer velocity does mean you don't have the same area scout watching a player develop over three or four years to get good context, but that's a small issue. Scouts will not lose track of players; they have constantly updated databases on transfer portals.

Ultimately, I think it's a bigger challenge for fans than for teams.

ArifMHasan · Wednesday at 3:41 PM

AlNipper49 said:
Hi Aarif,

Thabk you for coming here to talk to us.

I have a more basic question than most I presume. I have a 16 year old kid who is obsessed with following the NFL. It’s pretty clear you’ve taken a systemic approach to learning about the game. Other than your blog (which he’s not getting updates for) what other resources would you suggest for folks who want to follow the NFL and grow the knowledge of the game?

Basically, somewhere below becoming a Dan Hatman student and above an average WEEI listener

Books. The books I'd recommend that are approachable (in my opinion) and accurate (in my opinion)

Smart Football by Chris B. Brown
The Art of Smart Football by Chris B. Brown
Blood, Sweat & Chalk by Tim Layden
War Room by Michael Holley (for Patriots fans!)
More Than A Game by Ron Jaworski

Youtube channels that I'd recommend are Brett Kollmann's and Thinking Football's. Deeper down the rabbit hole might be Samuel Gold and AlexRollinsNFL.

Websites I'd recommend are the Ringer (Solak and Kelly are extremely approachable writers) and the Athletic (now hosted at the New York Times).

ArifMHasan · Wednesday at 3:50 PM

Ale Xander said:
Hi Arif,
Thanks for doing this. Do you think the NFL will be around in 30 years? Why or why not? What is the greatest obstacle, CTE awareness, greed, or something else?

Yes. I don't think CTE will be that big of an obstacle. Perhaps greed, though that's broadly defined. The NFL's desire to expand to new markets can create issues, especially as they think about possible league expansion. Expansion is convenient because 1) the rights to an expansion team are very expensive and that drops a lot of money into owner pockets immediately and 2) you've now captured the market of that expansion team. But it can create long-term problems both in terms of adding logistical and infrastructural challenges (flights, stadium negotiations, securing practice facilities, etc) and in diluting the product (there's only so much talent). I also think that "greed" could contribute to the NFL's increasing friction with local governments as they attempt to secure subsidies for stadiums and stadium improvements. The biggest issue in this category, however, is the challenge associated with ballooning franchise valuations. Buyers of teams are increasingly difficult to find, which means it will be tougher to kick an owner Snyder-style and deal with owner deaths (e.g. Walton-Penner buying the Broncos).

The number of potential buyers is dwindling and ownership groups are increasingly necessary. But even ownership groups can't raise enough capital under the current rule structures (there must be a limited number of people in the group and one ownership family must control at least 30 percent). So the NFL is considering the possibility of allowing a portion of a team to be owned by private equity; hedge funds like Blackrock could own 10 percent of teams. They may also consider the possibility of allowing foreign ownership, which raises the possibility of the Saudi royal family continuing to sportswash through franchise purchases. I think both private equity and Saudi royal ownership could diminish the product with short-term profit decisions (private equity) or extravagant non-sports decisions (House of Saud).

These do not strike me as overly large challenges, but they do strike me as potential pitfalls.

ArifMHasan · Wednesday at 3:56 PM

jacklamabe65 said:
Welcome, Arif!

Do you think the Patriots should sit down with Blake Maye and have him learn the ropes this year (as long as Jacoby remains healthy)?

Yea. Drake has a lot of things to address, in my opinion, and getting him more tightly controlled should be step one, if possible. I think his reckless playstyle is fine -- honestly, an asset -- but he needs to dial down some shots and take layups. I'd like some more accuracy. Generally, people say you can't improve accuracy in the NFL but that's not exactly true, in my opinion. If accuracy is an issue with inherent control, I don't think you can. If accuracy is an issue with the release, it's very difficult to improve. But if accuracy is a product of footwork, as appears to be the case for Maye, then it's much more fixable. He was not really coached on footwork in college -- this is fairly well-known -- so it's not as if he couldn't take to coaching on this point. If that can be hammered in, I'd try to do that before playing him.

ArifMHasan · Wednesday at 4:43 PM

Eck'sSneakyCheese said:
Welcome Arif! I think it's great that you're taking time out of your obviously busy schedule to lend some of your expertise to our conversations. I look forward to your replies on some great questions already brought up by others. I read a couple of the articles on your site and they were really informative.

I'm really interested in the draft evaluation process but unlike most here I'm leery of how quantitative something like this truly is. There are numerous variables at play that can't be measured. The human element and it's uncertainty is by far the biggest one. Knowing that there are multiple environmental attributes including the ever evolving landscape of the NFL, it's rules and it's personnel that can effect outcomes how can any of this number based analysis be trusted? Furthermore, taking into account the fact that there are so many people that have opinions on prospects now and some are more clued in than others are you worried about having so many boards creating noise in the final tallies?

Many of the most successful teams in the NFL deploy analytical insights (or align with them), so I'm fairly confident that there's value to it. The 49ers, Eagles, Ravens and Lions all tend to follow analytical principles in some respect -- though the Lions don't draft according to positional value, almost every other move they make follows those insights. The key I think is an open conversation between those who are conversant in data and those steeped in football.

I don't think the issue is the usage of numbers or forms of analytical analysis. A previous answer in this thread discussed the value of traditional scouting as data, and that's important. But the key is to not be wedded to some understandings of the data. A good example is fourth-down go-rates. For some time, analysts used success on fourth and short to argue that teams should go for it more often on fourth down instead of punting, particularly in short-yardage situations. Now, they were correct (to the degree that we can evaluate it), but they were overconfident. It turns out that fourth-and-short go-for-it success rates were a bad sample to use because they were poisoned by selection bias. Coaches were more often going to go for it on fourth down when they expected that they were going to succeed.

That could be because they were already kicking ass in the trenches or because 4th-and-1 in a box score can mean a lot of different things. If a ball is 1.49 yards away from the conversion line, it's treated the same in the play-by-play as if the ball were 0.05 yards away. But coaches making that call appreciated the difference and called it much more often in the latter situation than the former. Ball-tracking data verified this and found that there is a substantial reason fourth-down success rates are higher than third-down success rates at the same distance -- the ball is, on average, closer to the conversion line on fourth down when teams choose to go for it rather than on third down, when they don't have a (real) choice to do anything but go for it.

It's not a perfect example, because many contemporary analysts accounted for this by using mid-game third-down rates, etc. instead, but it was still instructive.

I highly recommend this piece, titled "Stop Thinking Like a GM, Think Like a Player" about the dangers of analytical insights being treated as gospel truth. It details two examples: a hypothetical example of a player with a high-rate of pop-up flies in balls batted in the field of play and catcher framing. In the first, the author details why a data analyst and a broadcaster might approach the same problem in two different ways (analyst: the pop-ups will regress over time, don't worry about it; broadcaster: he's letting his hands climb up too far up the bat) and both can be correct but the data analyst might scoff at the broadcaster and vice versa. Regression can occur because teams will correct their mistakes instead of merely being the product of just good or bad luck correcting itself.

In the second, the author details the story of a catcher who data analysts generally regarded as overrated because of his poor offense and had difficult understanding why he kept pulling in big contracts. After pitch framing data became well-known, the same analysts would argue the catcher was underrated because his ability to turn balls into strikes was so valuable that his poor offense didn't matter. And then they lauded themselves for promoting this catcher after he secured a huge contract!

It's a good read. I want you to know I am very familiar with this kind of skepticism of data analysis and I appreciate it. I think it's good. The best analysts incorporate their concerns into their approach.

ArifMHasan · Wednesday at 5:27 PM

sezwho said:
Thanks for joining! My question is a random one about dline play. With the continued trend towards multiple defensive fronts is it realistic/effective for a d tackle to go back and forth between 1 and 2 gap techniques even in the same series?

This is not as uncommon as you might believe. The only reason it's not that common is because two-gapping in general was dying, in part because stopping the run has become less important. Penetrating the pocket and pressuring the passer became paramount. But two-gapping can become an effective change-up. One interesting approach actually came from the Patriots; Bill Belichick would run a hybrid front with Vince Wilfork, where the offensive line was split in half at the center position, where one side would two-gap and the other would one-gap (this is detailed in one of the books recommended above, the Art of Smart Football). Often they would have to change who would do what, meaning that defensive tackles would have to be well-versed in both types of play.

Two years ago, the Vikings did the same thing with Michael Pierce at the nose. The defensive line coach publicly denied that they were hybrid but I was confident enough in what I saw to bring it to a few DL people and they confirmed that the Vikings actually did play both approaches.

More recently, the Rams changed their approach to "gap-and-a-half" type schemes to prioritize run defense. This, counterintuitively, improved their passing defense. The reason for that is because the beefed-up run defense of the defensive line allowed the Rams to lighten the box and deploy more players in coverage over a wider range of the field. It turns out that coverage controls those outcomes more than pressure and coverage creatures more pressure a bit more than pressure aids coverage.

I certainly think it's realistic and possibly effective, but only if you marry your front with your coverage.

ArifMHasan · Wednesday at 5:52 PM

dirtynine said:
Hi & thanks for dropping by.

Whenever I try to watch all-22 or even YouTube breakdowns, I get lost. Too many tiny moving pieces, not enough inherent understanding of what any particular offensive play is “supposed” to be or what the defense is “trying” to do. When you’re watching film, are there any basic pointers you’d give to inexperienced viewers that can lend a frame to things? A process or checklist to use before the snap to set up a mental model for what you’re about to see?

This is a bit more expansive than a Q&A can provide, but I think reading Blood, Sweat & Chalk by Tim Layden (recommended above) or the old New York Times Guide to NFL Defenses (archived links below) and then start by trying to identify the coverage shell by watching the safeties. Nothing else -- not if it's man or zone, not if there's a blitz, etc. Just see if you can tell if it's single-high (which can be Cover-1 or Cover-3), two-high (Cover-2 or Cover-4) or some hybrid (e.g. Cover-6). Once you feel comfortable with that, you can pay attention to what the corners do. I know the players are small, but you can get a sense of where their eyes are at the snap and how they move -- if the follow one receiver across the field, they're in man. If they stay rooted as a receiver leaves their area, they're in zone. See if you can get comfortable IDing man or zone corners. Some coverages will have both and there are more complicated things like pattern-match coverages that complicate it, but seeing if you can ID man or zone will help. And from there you can keep on going. I do think the Brett Kollman videos will help with that.

Same with offense. See if you can identify a route using the common route tree. And then see if you can identify two or three routes from receivers. From there, you can look up the tight end/slot receiver route tree (it's a bit different) and try to ID those. And grow from there. Once you get more comfortable, you can see what other elements of the offense you see in Blood, Sweat & Chalk that you can ID.

NYT Guide:

These were all published in 2010, so they are a decade behind. But the fundamentals are all there and they are well explained.

ArifMHasan · Wednesday at 6:12 PM

Bowser said:
Thanks, Arif, for spreading some wisdom.

Question: How seriously should we take any consensus board?

I assume we're referring to the consensus of talking head analysts and scouting services, i.e., not the teams themselves? Given that each team has a better understanding of its needs and how it values the various position groups, is there any real benefit in, say, critiquing the Pats' recent draft performance by comparing its internal assessment of needs (which we can't know) to some consensus of non-decision-makers?

I mean, it very much depends on how seriously you generally take outside opinions. As far as an outside opinion is concerned, it's pretty good! Above, I outlined some of the reasons I'm fairly confident in it and why it tends to match the NFL in overall predictiveness with regards to player performance -- and what happens when NFL teams reach against the board or have steals against the board. I'll let my response to the Top 30 Visits question and the response to the Top 100 prospects question serve as my response here.

I'll say this -- generally I've found that teams with low grades from the Consensus Big Board have very poor drafts relative to the rest of the league when we re-examine it 4 or 5 years later.

j-man · Wednesday at 6:21 PM

thanks for all this

ArifMHasan · Wednesday at 6:48 PM

Awesome Fossum said:
Oh great question. On a sort of similar theme: How do you watch football on TV? That is, when you sit down to watch a single game live on TV, what are you paying attention to pre-snap and post-snap? Are you just following the ball or something else? Are you taking mental notes? Physical notes!? Is there anything you'd recommend for people who want to be able to follow a given given at just a slightly more advanced level?

The books and YouTube channels I recommended above I think should give you a good headstart. Lately, I've stopped watching a single game at a time because it's become more important to have the pulse of the league -- I have four monitors going, one with the primary game I'm watching and another with redzone. One has a secondary game I'm interested in and the fourth has the rest of the games all on one screen. This absolutely sacrifices depth for me. I do not have the ability to watch games as deeply as before. But I can still note trends, storylines and overall feel. If a game interests me, I have no issue watching it back; I often have to watch stuff back anyway, often in All-22. I am by now pretty used to quickly figuring out coverage, routes, run scheme, etc so some of this is not a problem.

Generally speaking, what I brought up above about how to know what you're watching (e.g. ID the shell, then the CB coverage, etc) is what I pay attention to, though of course, you need to pay attention to Down & Distance, score and time remaining.

It's actually really difficult to break it down because some of it is second-nature at this point. If you ask a novice welder what they're paying attention to, they'll tell you they're focusing on stability, spacing, timing, gas flow, distance of the material to the surface, voltage, what material they're welding with etc. If you ask a veteran welder what they're paying attention to, many of them will say "the weld." They of course are paying attention to all of those other things -- they know that steel beads easier than aluminum if you ask them -- but they don't know that that's what they're doing unless you bring it up. And their welds look fucking perfect.

I'm not saying I watch games as well as a veteran welder welds (or as well as a coach watches games), but I've watched so many so often that the process has become natural. I noticed a pressure from a player who I don't recognize and I'm already opening a new tab with Ourlads half-typed before I realize what I'm doing. I see a deep touchdown and I automatically try to figure out what the safety rotation was supposed to be.

I think a good framework when you see something interesting, ask "whose fault is this," and "what could have stopped this from happening," and you find yourself learning a lot along the way.

dirtynine · Wednesday at 7:19 PM

ArifMHasan said:
This is a bit more expansive than a Q&A can provide, but I think reading Blood, Sweat & Chalk by Tim Layden (recommended above) or the old New York Times Guide to NFL Defenses (archived links below) and then start by trying to identify the coverage shell by watching the safeties. Nothing else -- not if it's man or zone, not if there's a blitz, etc. Just see if you can tell if it's single-high (which can be Cover-1 or Cover-3), two-high (Cover-2 or Cover-4) or some hybrid (e.g. Cover-6). Once you feel comfortable with that, you can pay attention to what the corners do. I know the players are small, but you can get a sense of where their eyes are at the snap and how they move -- if the follow one receiver across the field, they're in man. If they stay rooted as a receiver leaves their area, they're in zone. See if you can get comfortable IDing man or zone corners. Some coverages will have both and there are more complicated things like pattern-match coverages that complicate it, but seeing if you can ID man or zone will help. And from there you can keep on going. I do think the Brett Kollman videos will help with that.

Same with offense. See if you can identify a route using the common route tree. And then see if you can identify two or three routes from receivers. From there, you can look up the tight end/slot receiver route tree (it's a bit different) and try to ID those. And grow from there. Once you get more comfortable, you can see what other elements of the offense you see in Blood, Sweat & Chalk that you can ID.

NYT Guide:

Part 1: Introduction and Defensive Line Play

Part 2: Evolution of the 4-3 Front

Part 3: The 4-3 Front, Continued

Part 4: The 3-4 Front

Part 5: The Zone Blitz

Part 6: The 46 Defense

Part 7: Nickel Subpackages

These were all published in 2010, so they are a decade behind. But the fundamentals are all there and they are well explained.

Thank you Arif! This is a fantastic roadmap.

So cool of you to stop by. I’ll be checking out Wide Left.

ArifMHasan · Wednesday at 8:43 PM

I hope everyone is enjoying the schedule release!

Manuel Aristides said:
Hey Arif, thanks for being here.

One thing that I end up debating a lot with fellow football junkies is the health and dependability of running quarterbacks. Anecdotally, it feels easy to say that these guys get hurt more often. Richardson's rookie year felt like the archetypal running QB performance: inspired, game breaking talent when he's out there, but the same things that make him undefendable seem to increase his injury risk. While I personally feel like this is a "thing", some of my more data savvy friends have crunched numbers and found other numbers crunched by others and apparently the math is in opposition to my observations. Which, sure, I can believe I might be wrong: even just last year, significant injuries accrued even to obvious pocket-guys on all levels of the talent spectrum; Herbert, Cousins, Jimmy G, just off the top of my head.

The implications are obvious for both the fantasy obsessed (yo) and the actual gridiron. Teams seem to think there's not much to my fear, as running QBs continue to rise in drafts generally. Would love to know if you think there's an actual answer to the question "Is building around a running QB long term a bad idea, because they're more likely to be injured (or for some other reason not listed here, ie, maybe those guys never hit their ceiling as a passer or whatever)"?

Ideally would love your answer before my dynasty draft next week. (Kidding, mostly.)

I wish I could have a better answer here, but I think I have to defer to the data. There just isn't a great sample size or a good understanding of how we can demarcate running quarterbacks from scramblers. For example: is Russell Wilson a running quarterback? Was there a year where he suddenly wasn't one after being one? Many people don't think of him as one because there weren't that many designed runs in the offense for him, but there are two problems: 1) there were not many designed runs for Michael Vick for most of his career. His runs primarily came on scrambles. Is he not a running quarterback? 2) Russ earned 450 rushing yards or more in 5 of his first 6 seasons. In 2013, he had more rushing yards per game than Lamar Jackson had in 2023.

Was Daniel Jones always a running quarterback or only in 2022? Was Josh Allen a running quarterback in 2018, 2021 and 2022 but no other years? This is pretty important because injury studies can either restrict themselves to one-year data, which is not always what we're looking for, or attempt to determine long-term lifespan. I think we remember the notables -- like Robert Griffin -- and not the non-notables, like Lamar Jackson. Also, remember that Matthew Stafford was labeled as injury-prone early in his career and Luck had health problems too.

Generally speaking, mobile quarterbacks who last have done a great job of avoiding hits. I've noticed that both Lamar Jackson and Russell Wilson end their runs without being taken down to the ground, either going out of bounds or sliding. A lot of quarterbacks who have been injured take hits in the pocket. If we had a way of cataloguing all quarterback takedowns, separating tackles from runs out of bounds, etc., maybe we could do a better job but even then there are problems -- if you get to the second-level consistently, the hits you're taking are from DBs, not DTs. But they are moving a bit faster, so maybe that hurts more than helps? I don't know.

Honestly, just trust the reams of data suggesting that mobile quarterbacks are safer than non-mobile QBs.

ArifMHasan · Wednesday at 11:27 PM

SMU_Sox said:
Arif,

Thank you for coming on board!

Other people have covered the Patriots draft questions but I’ll add that it looks like aside from Drake Maye the rest of their draft was heavy negative expected ROI%(according to your piece given Maye being one of the best values but a -2.2% expected ROI overall for the whole draft) largely because they reached for Wallace, Polk, and then Robinson as well. Would it be fair to give them, and this is how I see it, an A for Drake, a C-/D+ for the rest of the draft and a B-/C+ overall?

If I were really strict about the normal bell distribution of grades and applied ROI% to a bell curve for that purpose, the Drake Maye pick would earn a B in the system, but percentage value gained to letter grade doesn't really work well in this application because it is impossible to pick in the first round and get an A. The highest grade given in the first round this year would have been a B. If I re-run it for a distribution inside each round, the Maye pick gets... a B. The A's go to Marvin Harrison to the Cardinals, Quinyon Mitchell to the Eagles and Terrion Arnold to the Lions. The Maye pick is regarded as the sixth-best grade in the round, so take that for what it's worth.


Player	Round	Overall Grade	Round-Adjusted Grade
Drake Maye	1	B	B
Ja'Lynn Polk	2	C	D+
Caedan Wallace	3	D	F
Layden Robinson	4	D	D
Javon Baker	4	B	B
Marcellas Dial	6	D	C-
Joe Milton III	6	B	B
Jaheim Bell	7	B	B-

By ROI% the team lands in the middle -- ranking 14th overall. That would give them a C.

Personally, I'd say the overall grade is fine, but I'd give them an A for the Maye pick. Sure, why not.

SMU_Sox said:
My question: You have discussed on X/Twitter about how younger players can play learn physical tools easier than older players. I believe it was a neurological explanation. This makes for draft projection implications. The Patriots under Bill traditionally skewed older and this draft from Wolf also skewed older (like say drafting a 24 year old OL pick 68 or a senior WR at pick 37 both moves I didn't like for where they were on the board and not just because of the age). Can you tell the board a little bit about the science behind this? Any decent links?

Edit: One of the biggest issues I had with Wallace was, like Cole Strange who was also an older prospect coming out, he is primarily a two-hand-striker in pass pro. If I think back to the three guys with the worst independent UoH (use of hands) coming out in the top 75 recently it's anecdotally Cole Strange, Liam Eichenberg, and Caedan Wallace. It's another crucial skill in pass pro. I wonder how difficult it will be for older players vs younger players to develop this trait.

As for older/younger players when it comes to offensive line development, my fallback is to think about research into neuroplasticity. As it relates to motor function, there's not a ton of research on early adulthood but what I can find suggests that it declines a little bit after age 25. This is for a general population and does not include professional athletes, who tend to have elite cognitive capacity but also suffer from brain injury at a higher rate. I also think about players drafted over 25 years old -- people like Brandon Weeden (28-year-old quarterback, finished the year age 29) and Danny Watkins (26-year-old offensive lineman, finished the year age 27). It's not a great track record but there are some standouts among older players who were drafted -- Garrett Bolles, Bruce Irvin, Kyle Long, etc. I think Long isn't a good comparison because I recall he was supposed to be refined coming out. But Bolles wasn't and he was drafted at 24, finishing the year age 25.

When it comes to rewiring your brain so that your hands can act independently, I'm not confident.

Here's a study that has charts starting at age 25 and this one with charts on fluid and crystallized intelligence that suggest that it peaks at age 26 and begins declining. They worked off of this well-cited work in 2008 that began exploring the question seriously in 2008.

This is all conjecture on my part. I am not a scientist, I am not well-versed in neuroscience, cognitive science or neurobiology. What I know is that players who are older tend to underperform their similarly-drafted teammates. I know that players who produce at a remarkably high level in college at an older age tend to avoid this (Keyshawn Johnson is a great example of this, he accounted for an astounding 43.6 percent of his team's receiving yards). Essentially, think of it like a sliding scale -- the older you are, the better you need to be in order for you to justify your spot over a younger player. I know PFF grades aren't everything, but they are suggestive. And it's difficult for me to believe Caedan Wallace blew his competition out of the water while still only managing a 68.8 grade.

EDIT: Edited to fix typo with Javon Baker's grade

ArifMHasan · Wednesday at 11:48 PM

SMU_Sox said:
I'll add one last question that has been tough for me to solve. How should I or anyone who creates big boards positionally rank them (I'll add that your answer to this should help us think about how GMs might want to positionally rank theirs as well)? I have been taking my grade and then adding 1 point for QBs. For non QBs I take their grade and then add a fractional point to theirs. To get the fractional point I take the average top 15 salary for each position divided by the average top 15 QB salary. So if the top WRs on average make 65% of the top QBs I would add 0.65 points to my WR score (made up that number). Total scale is 9 points. Then for the guys I do not project as starters I divide all those points by 2. I think this appropriately weighs QBs but can overrate WRs for example. I also don't split guys up as X's, slots, Z's, etc.

Honestly, this is something I've been trying to wrestle with for a bit and I'm not entirely sure. I don't get pure talent boards for the Consensus Board, I get the final versions of boards after people have already done positional weighting. Historically, I've adjusted the CBB to the NFL draft for grading purposes by giving them a historical positional weight that makes ten year's worth of boards match ten year's worth of drafts. But I think a system that leverages off of historical draft position or historical salaries works. I think one thing you should do to avoid the overweighting problem is take averages by proportional snap count. So average of the top 5 QBs (QBs took 35,759 snaps last year) would be comparable to the average of 13 WRs (WRs took 92,668 snaps last year -- 35,759 divided by 92,668 is 0.3859, 5 divided by 0.3859 is 12.96). That means for every QB snap, there are 2.59 receiver snaps, which matches our intuitive expectations.

I may work on this in the off-season.

Manuel Aristides · Thursday at 7:03 AM

ArifMHasan said:
Honestly, just trust the reams of data suggesting that mobile quarterbacks are safer than non-mobile QBs.

Thanks, Arif. I don't know why this one is difficult for me -- generally I'm the sports fan insisting my friends believe the data. Some ingrained bias here, apparently. Appreciate your perspective.

Eck'sSneakyCheese · Thursday at 11:01 AM

ArifMHasan said:
Many of the most successful teams in the NFL deploy analytical insights (or align with them), so I'm fairly confident that there's value to it. The 49ers, Eagles, Ravens and Lions all tend to follow analytical principles in some respect -- though the Lions don't draft according to positional value, almost every other move they make follows those insights. The key I think is an open conversation between those who are conversant in data and those steeped in football.

I don't think the issue is the usage of numbers or forms of analytical analysis. A previous answer in this thread discussed the value of traditional scouting as data, and that's important. But the key is to not be wedded to some understandings of the data. A good example is fourth-down go-rates. For some time, analysts used success on fourth and short to argue that teams should go for it more often on fourth down instead of punting, particularly in short-yardage situations. Now, they were correct (to the degree that we can evaluate it), but they were overconfident. It turns out that fourth-and-short go-for-it success rates were a bad sample to use because they were poisoned by selection bias. Coaches were more often going to go for it on fourth down when they expected that they were going to succeed.

That could be because they were already kicking ass in the trenches or because 4th-and-1 in a box score can mean a lot of different things. If a ball is 1.49 yards away from the conversion line, it's treated the same in the play-by-play as if the ball were 0.05 yards away. But coaches making that call appreciated the difference and called it much more often in the latter situation than the former. Ball-tracking data verified this and found that there is a substantial reason fourth-down success rates are higher than third-down success rates at the same distance -- the ball is, on average, closer to the conversion line on fourth down when teams choose to go for it rather than on third down, when they don't have a (real) choice to do anything but go for it.

It's not a perfect example, because many contemporary analysts accounted for this by using mid-game third-down rates, etc. instead, but it was still instructive.

I highly recommend this piece, titled "Stop Thinking Like a GM, Think Like a Player" about the dangers of analytical insights being treated as gospel truth. It details two examples: a hypothetical example of a player with a high-rate of pop-up flies in balls batted in the field of play and catcher framing. In the first, the author details why a data analyst and a broadcaster might approach the same problem in two different ways (analyst: the pop-ups will regress over time, don't worry about it; broadcaster: he's letting his hands climb up too far up the bat) and both can be correct but the data analyst might scoff at the broadcaster and vice versa. Regression can occur because teams will correct their mistakes instead of merely being the product of just good or bad luck correcting itself.

In the second, the author details the story of a catcher who data analysts generally regarded as overrated because of his poor offense and had difficult understanding why he kept pulling in big contracts. After pitch framing data became well-known, the same analysts would argue the catcher was underrated because his ability to turn balls into strikes was so valuable that his poor offense didn't matter. And then they lauded themselves for promoting this catcher after he secured a huge contract!

It's a good read. I want you to know I am very familiar with this kind of skepticism of data analysis and I appreciate it. I think it's good. The best analysts incorporate their concerns into their approach.

Thanks Arif. This thread has been really informative. The Molina story was great and reminded me a little of the back and forths we had here about Christian Vazquez. Sabermetrics is what brought me to love then hate data analysis. Then PFF really got me to question how useful it is in the NFL. (I'm not a fan.) As someone who used to fap to FIP I do find validity in most of it but I have a hard time when people try to push definitive outcomes based off of this information. Like Vazquez would never in his career hit over .200 (said by some here.)

I liked your response to Dotrat and felt that answered some of my questions about your overall process. I appreciate you really trying to take in all factors but I still question the number of boards. I have doubts that 100+ different data points are all relevant. I'm interested in reading more about Eric Eagers process as well as the PCA approach you mentioned.

This brings me to Caedan Wallace. What a polarizing pick. Graded everywhere from a B to an F. You too seem to be critical of the pick. His 10 yd split, broad and vert were marked elite. Traits that you've said are important, especially for OL. Long reach. Powerful punch. Experience practicing on the left side his entire college career, also played LT all throughout High School. Showed up well at the Shrine Bowl playing both guard and LT. Multiple top 30 visits as mentioned above. Coming into a situation where I believe Scott Peters (underrated hire IMO) is going to put him in a position to succeed. Yet, it's being extrapolated that the ROI is going to be poor. How? Because some peoples perception of the data differs? I think it can be easily parsed that the Pats had him as a possible LT while other teams had him solely on the right which made for the discrepancy.

Also, a B for overall grade for Baker but a round adjusted grade of a D? Can you explain this a little more?

Eck'sSneakyCheese · Thursday at 1:00 PM

SMU_Sox said:
Coach Williams (mightkaos) on twitter used Chat GPT to do what Eager did but just for corners. Was a super interesting experiment.

I think for Wallace he is generally more of a negatively graded pick vs polarizing. Or, to use metacritic, mixed reviews mostly negative. The Shrine Bowl is nice but is usually considered a day 3 prospect or UDFA bowl. Aside from Zay Flowers the best prospects go to the Senior Bowl traditionally and do not go to the Shrine Bowl. Practicing at LT or doing ok there in the all-star circuit is fine but not really indicative of actual prowess. There are so many things that are different in the all-star circuit. You don't have to process as much for one because the defense is limited in what scheme they can play and what their play calls are. That's huge.

As for his actual eval, I am not sure if Arif watched him or not, but he doesn't have good independent use of hands and is primarily a 2-hand puncher type. He's a chronic over-setter. He gets a lot of support with inside counters. He doesn't gain enough depth in his kickslide. He consistently cheats by opening his hips up early too. He lacks elite footspeed. These are some of the reasons why Brandon Thorn thought he might have to kick inside to guard. Kicking inside to guard vs being on LT island is a huge jump.

As for the number of the boards - I'm curious why this is an issue. When we think about the wisdom of the crowds in general we know that there are some idiots in the crowd but overall it has merit. And if the CBB is so accurate at predicting various outcomes then why does the number of boards in it matter?

Edit: not saying I wouldn't ask the question or anything but thinking about if it matters or not and if it matters why it matters. You aren't the only one who has brought this up.

Mixed reviews mostly negative is a fair assessment. I chose polarizing because I saw the reviews more as equally positive to negative. Theres some variance there and I'm sure you've read more than I have.

I disagree that practice and B-List All Start game performance are not indicative of ability. How else are we or anyone supposed to judge? The fact he wasn't a complete dumpster fire says something. Maybe not to any definitive degree but being able to do the job well in theory (practice) should account for something.

The issues he has, that you listed a few of, I believe are coachable. Other than foot speed obviously. If he had better agility numbers I don't think he would have been available at 68. You're saying kicking inside vs LT is a huge jump. Why are those the only options? It's less of a jump going from RT to LT which is actually what he's doing.

For me the number of data points comes down to quality over quantity. Especially now where the negatives can take over fairly easily due to negative views being the ones people gravitate towards. If you're just adding data to add data you're introducing noise. At what point do the views start to muddle due to bias, misinformation and poor judgement? How do you even know those are factors? Does JimBobs Texas Sized Big Board actually have any value? (not a real board, I hope, but you get what I mean.)

Arif Hasan Q&A

queer eye for the next pats guy

New Member

Member

Hamilton

queer eye for the next pats guy

Hamilton

Joe wants Trump to burn

queer eye for the next pats guy

Yukon Cornelius

Member

Well-Known Member

New Member

Huge Member

Hamilton

Well-Known Member

Member

Member

Member

New Member

Member

Member

Member

Member

Member

Member

Member

Member

queer eye for the next pats guy

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

queer eye for the next pats guy

Member

Member

Member

Member

Member

Member

queer eye for the next pats guy

Member

queer eye for the next pats guy